Chapter 2

Architecture Landscape

By Mårten Ericson, Bahare Masood Khorsandi, et al.1

Copyright © 2023 Mårten Ericson, et al.

DOI: 10.1561/9781638282396.ch2

The work will be available online open access and governed by the Creative Commons “Attribution-Non
Commercial” License (CC BY-NC), according to https://creativecommons.org/licenses/by-nc/4.0/

Published in Towards Sustainable and Trustworthy 6G: Challenges, Enablers, and Architectural Design by Ömer
Bulakçı, Xi Li, Marco Gramaglia, Anastasius Gavras, Mikko Uusitalo, Patrik Rugeland and Mauro Boldi (eds.).
2023. ISBN 978-1-63828-238-9. E-ISBN 978-1-63828-239-6.

Suggested citation: Mårten Ericson, Bahare Masood Khorsandi, et al. 2023. “Architecture Landscape” in Towards
Sustainable and Trustworthy 6G: Challenges, Enablers, and Architectural Design. Edited by Ömer Bulakçı, Xi
Li, Marco Gramaglia, Anastasius Gavras, Mikko Uusitalo, Patrik Rugeland and Mauro Boldi. pp. 11–39.
Now Publishers. DOI: 10.1561/9781638282396.ch2.

1. The full list of chapter authors is provided in the Contributing Authors section of the book.
2.1 Introduction

The network architecture evolution journey will carry on in the years ahead, driv-
ing a large scale adoption of 5th Generation (5G) and 5G-Advanced use cases with
significantly decreased deployment and operational costs, and enabling new and
innovative use-case-driven solutions towards 6th Generation (6G) with higher eco-
nomic and societal values. The goal of this chapter, thus, is to present the envisioned
societal impact, use cases and the End-to-End (E2E) 6G architecture. The E2E 6G
architecture includes summarization of the various technical enablers as well as the
system and functional views of the architecture.
The design of the 6G architecture is based on the analysis of the societal, eco-
nomic, regulatory, and technological trends, which are discussed in Section 2.1.2.
A summary of the use cases envisioned for 6G is also introduced in Section 2.1.3.
Accordingly, a set of architectural principles has been drawn, upon which the pre-
sented architecture is built. Herein, the main highlights of the 6G system design
are provided, while the details on the various network domains are given in the
subsequent chapters.

11
12 Architecture Landscape

In Section 2.2, the overall architecture description discusses the new stake-
holders in the mobile network ecosystem, and how the architectural work is tak-
ing into account their requirements in all the domains of the network. Specific
design principles that need to be factored in for the new architecture are also
described. Section 2.3 discusses the components of the security architecture, which
are required and must be applied to have security as a design principle for the 6G
architecture. A deep dive into the Management and Orchestration (M&O) archi-
tecture is then presented in Section 2.4. Section 2.5 outlines the summary of this
chapter and presents the outlook.

2.1.1 The Societal Impact of 6G

Since the invention of mobile telephony half a century ago, wireless network tech-
nology has undoubtedly transformed the everyday lives of billions of people on the
planet, and profoundly shaped the economy and the evolution of human society
to date. The mobility of communication and of access to information has allowed
completely new ways of interacting, working, and evolving our communities. For
each generation of wireless technology, the applications and usages have become
increasingly ingrained in our societies and have become an established backdrop
to our modern lives. 6G will continue to impact our societies and will enable that
communication is always possible and information is always available.

2.1.2 Trends and Evolution Towards 6G

Today, when the world is facing several unprecedented challenges in parallel and
the prosperity of human society and the long-term survival of mankind are in peril,
access to information and the possibility to communicate everywhere are a must.
From climate change to global pandemics, social inequalities, misinformation, and
distrust in democracy, addressing any challenge that impacts today’s global eco-
nomic, societal, and political agendas requires further and sustainable digitalization
of the global economy and society. Infused by emerging and disruptive digital tech-
nologies on the horizon, wireless networks are and will be the keystone for enabling
such a transformation. The network evolution during this and the next decade will
enable a large scale adoption of use cases to sustainably combat our challenges and
enable higher economic and societal values at a significantly decreased operational
cost.
As the Internet revolution played out over the past decades, with mobile broad-
band altering our interactions, professions, and habits in unforeseen ways, the true
social impact of 6G can only be ascertained in hindsight. Nevertheless, the kernel
of its potential can be considered through the current societal and economic trends
Introduction 13

towards 2030 and beyond, which will be analysed in the following sections. In addi-
tion, regulatory and technological trends that are critical for the design and deploy-
ment of future networks will be discussed, ensuring the vision and the research work
encompass all the essential elements and will lead to a future network design that is
deeply rooted in reality and profoundly benefits humanity in the mid-to-long term.

Societal trends towards 2030 and beyond

In 2015, 17 interlinked Sustainable Development Goals (SDGs) were collectively
identified by the General Assembly of the United Nations (UN) [1]. Since then,
all sectors of society have been called for working towards and delivering on these
goals with a timeframe of 2030 and beyond. The Information and Communica-
tion Technology (ICT) and wireless network industries have positively contributed
to many of the goal areas so far (e.g., to combat poverty and CO2 emissions), and
the potential of further contributing and successfully progressing towards the goals
is huge. In developing future networks towards 2030, there is a consensus among
major stakeholders from industry, academia, and policy makers around the world:
network technology shall support and further accelerate this change for a better
and sustainable world, and the network industry will increase its share of contri-
butions and responsibilities to society, enabling significantly increased efficiency in
the use of resources and facilitating new and sustainable ways of living in the next
decades [2–9].

Economic trends towards 2030 and beyond

Wireless network technology has long been regarded as an important engine for
driving global economic growth. As projected in [10], network technology that
encompasses 5G and beyond will potentially trigger $13.2 trillion in global sales
across ICT industry sectors by 2035, representing 5% of global real Gross Domes-
tic Product (GDP), while 6G value chain will be able to generate 22.3 million
jobs globally by 2035. This estimation did not even include the impact of con-
nectivity on non-ICT sectors. As recognized in [11], “the next era of industry will
be one where the physical, digital and human worlds are coming together,” facing
great economic and societal challenges towards 2030. Future networks will be a key
enabler for such a revolution with advanced technological capability and human-
centric design. New use cases will offer new growth opportunities assuming exist-
ing business models, and they can also drive and inspire new business models in an
evolving revenue ecosystem.

Regulatory trends towards 2030 and beyond

While the telecommunications sector has been liberalized and privatized in the
1990s, sector regulation continues to be important in conjunction with efficient
14 Architecture Landscape

spectrum access rules, aspects of electromagnetic field (EMF), and assurance of

level playing field with platform and cloud operators beyond telco context. Towards
2030, this trend will continue. For example, spectrum management is at the heart
of future networks and any wireless technology development, and governments and
regulators will have new opportunities due to a wide variety of spectrum bands with
highly distinct deployment characteristics and spectrum access models with differ-
ent levels and needs of spectrum sharing. Another rising issue is EMF exposure.
The deployment of 5G technology has started in different areas of the world, and in
some regions (including Europe), concerns over EMF exposure fuel the opposition
of the public to its rollout [12, 13]. The exposure to EMF is and will be regulated,
based on guidelines from the International Commission on Non-Ionizing Radia-
tion Protection (ICNIRP) [14]. Since the beginning of telephony, regulations have
played an important role in shaping innovation and the operation of the telecom-
munications industry, for example, setting the industry to be monopolies in the
1960s, liberalizing the sector with privatization in the 1990s, and setting up new
regulations for 5G local and private networks. Future networks will likely com-
bine a range of radio access network (RAN) technologies from macro cells to small
cells with very high-capacity short-range links. This calls for refining regulations
to resolve inconsistent local approval processes and frequency band assignments to
enable dense small cell deployments.

Technological trends towards 2030 and beyond

Previous generations of mobile networks have incessantly increased the perfor-
mance and capacity to connect and communicate, facilitating global mobile com-
munication and an intertwined global economy accessible at our fingertips. The
efficiency gains provided to industries facilitate complex global logistics chains, and
the interactive media have engendered a plethora of novel services and industries.
This trend is not foreseen to abate in the foreseeable future but is rather expected
to expand and encompass even further aspects of our societies.

Technology evolution towards cost reduction and improved efficiency

• Reimagined network architecture: As the applications grow more demand-
ing and diverse, the complexity of the deployments and management of the
system increases, and the possibility to flexibly and dynamically scale and
control the network resources in an efficient way becomes more important.
In 5G, the core network (CN) was reimagined into a service-based architec-
ture (SBA), which leverages the virtualization of network functions to only
instantiate the functions needed at each instance. Current trends in network
architecture point to an extension of the SBA further out into the RAN
allowing more flexible and autonomous operation of the network. A new
Introduction 15

network architecture paradigm for the 6G era is driven by a decomposition

of the architecture into platform, functions, orchestration, and specialization
aspects [15]. Future network platform will be associated with an open, scal-
able, elastic, and agnostic heterogeneous cloud, which is data-flow centric and
will include hardware acceleration options. Functionally, the convergence of
RAN and CNs will help reduce architectural complexity. At the same time,
options of flexible offload, extreme slicing, and flexible instantiation of sub-
networks will drive the increased level of specialization of the architecture.
Of high relevance for the open provision of services and the monetization
of resources will be the transformation of orchestration architecture; cogni-
tive closed loops and automation are likely to become pervasive. All future
deployment scenarios will rely on a superior transport network and network
fabric that is flexible, scalable, and reliable to support demanding use cases
and novel deployment options, such as a mixture of distributed RAN and cen-
tralized/cloud RAN enabled by AI-powered programmability [2]. The net-
work architecture shall provide the capability to facilitate all the AI operations
in the network.
• Improved network capacity: The ever-increasing demand for network
capacity necessitates the provision of additional bandwidth. The potential
to utilize the higher frequency bands, such as the sub-THz (100–300 GHz)
range, is currently being explored. However, the radio propagation is signifi-
cantly attenuated at these frequencies, and the reduced diffraction makes the
connection more susceptible to blockage. Coupled with the reduced power
efficiency and increased noise at higher frequencies, this compounds to sev-
eral technological challenges that need to be overcome to provide sufficient
coverage.
• New devices and interfaces: Future networks will be connected to multi-
tudes of devices and interfaces beyond mobile phones or computers, enabling
novel human–machine/machine–machine. New human–machine interfaces
created by a collection of multiple local devices will be able to act in uni-
son [3]. In addition, the ubiquity and longevity of Internet of Things (IoT)
devices will be further enhanced through zero-cost and zero-energy devices
where printable, energy harvesting devices can be deployed anywhere.
• Network of networks: In order to capture local and specialized network and
sub-network needs, 6G network-of-networks will cover multiple scales of –
physical and virtual – networks. The evolution of private and 5G non-public
network (NPN), such as campus networks, will expand to support many
machines and process with strict requirements on quality of service (QoS)
and connectivity, employing edge processing for further automation. With
16 Architecture Landscape

digital twins (DTs), massive data harvesting from local sensors builds up cap-
illary sub-networks handled by gateways, while in parallel the wide-area net-
work must handle mobility and coverage.
• Trustworthy networks: As more and more aspects of our lives, societies, and
industries become reliant on mobile connectivity, it becomes imperative to
ensure the performance, reliability, and security of the networks so that the
services can be used as intended, when needed, without undue connection
disturbances or access to private data. This will require the network archi-
tecture design to consider the security implications at every step, to avoid a
patchwork of solutions after the fact.
• Sustainable 6G and 6G for sustainability: As one of the major challenges
facing our societies today, the sustainability of our environment, industries,
and the society at large must be ensured to be able to reach the sustain-
able development goals set by, e.g., the United Nations. For 6G, this entails
addressing both the first-order effect of the network, referring to the direct
environmental impacts of the manufacturing and operating of the networks
in terms of energy consumption, CO2 emission and usage of scarce resources,
as well as the second-order effect, referring to how the networks enable
improvements in sustainability with, e.g., improved efficiency in industries,
or a transition from business travels towards virtual business meetings. How-
ever, there are also higher-order effects, also known as rebound effects, that
must be considered, where the improved functionality of the mobile networks
induces a novel behaviour of the users, which could, e.g., increase the total
energy consumption. Moreover, societal sustainability should be addressed,
with new services enabled by 6G meeting societal needs and demands [16].

Disruptive technologies that will shape future connectivity

• Convergence of communications, localization, imaging, and sensing:
With the use of wider bandwidth signals coupled with high band spec-
trum (>100 GHz) as well as the incorporation of simultaneous local-
ization and mapping (SLAM) with communications at lower frequencies,
future networks will be designed integrating high-precision localization (with
centimetre-level accuracy), sensing (both radar-like and non-radar-like), and
imaging (at millimetre-level) capabilities. This requires the development of
highly novel approaches and algorithms to co-optimize communications,
sensing, and/or localization.
• Network intelligence: The evolution of artificial intelligence (AI)/machine
learning (ML) has progressed in the past decades and may bring major disrup-
tions to future networks. Their applications are currently designed for specific
tasks, but as the development progresses, more general-purpose applications
Introduction 17

emerge. As this development occurs in parallel with and in conjunction with

the development of the mobile networks, it is foreseen that there may be
several synergies between them. By leveraging on the mobile access, the AI
agents can operate in a distributed fashion, gathering, analysing, and acting
upon data available across different localities on a much larger scale. At the
same time, the AI functionality can be utilized to optimize and enhance the
network operations to improve the performance and reduce the operating
costs by impacting the design of air interface, data processing, network archi-
tecture, and management towards computing for achieving superior perfor-
mance [3, 7]. It will become essential for the E2E network automation dealing
with the complexity of orchestration across multiple network domains and
layers [15]. This may also bring forth fundamental changes in how the mobile
networks operate, when there are AI agents both managing and operating the
networks, as well as transmitting and receiving the information being com-
municated, and the fundamental tenets of the network architecture design
may need to be revisited. For instance, the AI agents may be able to opti-
mize the network behaviour in near real time, which would necessitate the
ability to reprogram the network functions. This programmability would go
beyond the configurability available today and would allow the modification
and introduction of novel functionality into already deployed equipment.
• Digital twin: A DT is a digital replica of a living or non-living entity, physical
object, or process. The virtual representation reflects all the relevant dynam-
ics, characteristics, critical components, and important properties of an orig-
inal physical object throughout its life cycle. The creation and update of DTs
relies on timely and reliable multi-sense wireless sensing (telemetry), while the
cyber–physical interaction relies on timely and reliable wireless control [17]
over many interaction points where wireless devices are embedded.

2.1.3 Use Cases: Revolution or Evolution?

In previous generations of mobile networks, the use cases were straightforward:
how to provide voice and, later, data communication with increasing bit rates to
mobile devices. For the 5G, the use cases were broadened beyond enhanced mobile
broadband (eMBB) to include massive machine type communication (mMTC),
requiring low data rates with very low power consumption to enable connectivity to
billions of simple devices, as well as critical machine type communication (cMTC),
later termed ultra-reliable low-latency communication (URLLC), instead requiring
extreme reliability and low latencies [18].
With 5G Advanced, extended reality (XR) has been introduced as a prominent
use case, which encompasses both virtual reality (VR) and augmented reality (AR).
18 Architecture Landscape

In XR, different on-body or head-mounted devices can be used to experience the

digital world either fully immersed (VR) or overlaid onto the physical world (AR).
With the recent development and deployment of the 5G networks, the current
needs in the developed markets appear to be satiable with current technology at
least at peak performance, primarily necessitating a network densification to meet
the capacity and latency demands.
In 6G, it is foreseen that the incumbent use cases will continue to be preva-
lent, with access to mobile broadband extended even further, by incorporating non-
terrestrial networks and satellites to cost-efficiently reach remote and underserved
areas.
Considering again the example of XR use cases, this XR use case is projected
to increase in relevance as the devices improve in performance and form factor,
and improvements in the network can transition this use case from stationary use
near a hotspot towards mobile outdoor use. To ensure the performance in terms
of data throughput and low latency, while maintaining the small form factor, it is
expected that the XR use case will have to leverage on computational offloading,
where significant amounts of the data processing are completed in the network
instead of on the device. Similar to this XR example, other 5G use cases will evolve
towards extended range of usage, thanks to the developed capabilities offered by
6G, reaching more people and devices and allowing for usage in more extreme
conditions.
When it comes to revolutionary use cases, the introduction of novel function-
ality, such as joint communication and sensing, integrated AI functionality, or
energy-neutral devices relying on ambient energy harvesting, could enable unfore-
seen usages and applications of the future 6G networks. The development of “6G
for sustainability” use cases, in collaboration with different sectors and verticals,
could also open the way to revolutionary usages. Relying on 6G as a tool to con-
tribute to the reduction of the environmental footprint of other sectors could
indeed lead to new roles for 6G and mobile networks, beyond the traditional mar-
ket of previous generations. Although the International Telecommunication Union
Radiocommunication Sector (ITU-R) is working on defining the usage scenarios
for 6G, a few possible extensions to the previous usage scenarios can be envisioned,
e.g., further enhanced mobile broadband (FeMBB), ultra-massive machine-type
communications (umMTC), extremely reliable and low-latency communications
(eURLLC), long-distance and high-mobility communications (LDHMC), and
extremely low-power communications (ELPC) [19, 20].

6G use cases
The societal and economic trends are driving the identification of relevant use cases
for 6G. The Hexa-X project provides a vision on the role of 6G in the evolution
Introduction 19

Telepresence Sustainability

Robots 6G Resilience

Twinning Trust

Figure 2.1. Generic use-case families for 6G.

of society [21–23], accounting for these trends and setting the baseline for the
identification of use cases. Combining the sets of use cases identified by the vari-
ous European projects provides an overview of the envisaged usage enabled by 6G.
These different use cases can be clustered into broad and generic use case fami-
lies, encompassing both evolutionary use cases and revolutionary ones, building
on new functionalities. These generic use case families can be considered from
the perspective of the type of end-user usage involved, as shown in Figure 2.1,
such as:
• Telepresence: Immersive experience is a central theme for various use cases,
with different degrees of immersion, from the evolution of XR experienced
with 5G but with increased mobility, reliability, and scale to a fully immersive
experience, fully merging physical and digital worlds, with various application
areas, both professional and personal (travel, gaming, sports, etc.). This will
leverage both the expansions and evolutions of existing technologies, provid-
ing connectivity as well as incorporating novel functionalities such as local-
ization, sensing, and computational offloading.
• Robots: In parallel with the development of 6G, the evolution of robots and
autonomous systems will continue, and robots are envisioned to become part
of everyday life, both in professional and personal settings. They will collabo-
rate and interact with each other, but also with humans. The generalization of
robots will increase productivity but will also offer solutions to assist humans
in their daily lives, meeting societal demands such as care of disabled persons.
Although many aspects of this use case may be served by existing technolo-
gies, the increased demands for concurrent reliability, high bitrate, and low
latency necessitate novel approaches.
• Twinning: The concept of DT will be extended in 6G, generalizing the
use of the full digital representation of an environment to enhance control,
20 Architecture Landscape

management, and maintenance of different flows and objects to various

activity sectors. To capture, store, analyse, and distribute the digital repre-
sentation of the environment, it will require a seamless network of unprece-
dented scale, incorporating sensing, computational offloading, and connec-
tivity with low latency to numerous devices at the same time.
The generalization of these new services will also call for a new generation with
increased capabilities to support large deployments.
Other use-case families can be identified according to the research challenges
and values addressed:
• Sustainability: 6G can be a solution, for various verticals, to enable new
use cases contributing to the reduction of their environmental impact
(agriculture, industry, logistics, smart city, etc.). 6G can also help meet-
ing societal demands by facilitating access to key institutions and enforc-
ing human rights, such as access to healthcare, education, and reduction of
inequalities.
• Resilience: Various 6G use cases are built upon resilient infrastructure, guar-
anteeing the delivery and quality of service despite the complexity of the net-
work and possible situations and events. A resilient 6G network can be an
asset to improve and develop key usages (e.g., in the automotive sector) or to
develop new usages (e.g., facilitating public protection).
• Trust: A high level of trust in 6G networks is a prerequisite to various use
cases, involving sensitive information or operations.
Other use-case families can be identified, related to capabilities offered by the
network, either related to the management and operation of the network. New use
cases can also be enabled, thanks to new capabilities introduced by 6G, such as
sensing, positioning, AI processing, or compute capabilities.
Each use-case family encompasses a wide range of usages, from evolutionary
ones, extending and enriching the 5G usages with new capabilities, to more disrup-
tive ones, opening up new horizons where 6G could benefit and transform society.
6G use cases can also be evolutionary, relying on improvement of existing technolo-
gies, but also revolutionary when introducing new capabilities, such as sensing, AI,
and compute capabilities as well as novel devices and interfaces.

6G requirements
Like the use cases, requirements for 6G can also be categorized into the evolution
of key performance indicators (KPIs), e.g., higher throughput, lower latency, and
revolution of novel KPIs (Figure 2.2). These novel KPIs explicitly focus on the
E2E view required by novel 6G use cases, such as E2E dependability or service
The Need for a New Architecture 21

New capabilities

Localization and
Communication AI and Computation ...
Sensing
Availability

Dependability
Reliability

Safety
QoS Attributes

Integrity

QoS Attributes

QoS Attributes
Maintainability
Service latency AI Service RTT Location/orientation accuracy

Data rate Inferencing accuracy Sensing service RTT

Domain-specific
Resource constraints Interpretability level Refresh rate QoS attributes

Scalability Resource constraints Resolvable range

... Scalability Angular/velocity resolution

... Resource constraints General QoS

attributes
Scalability

...

Figure 2.2. Classification of 6G KPIs.

availability. With the envisioned novel capabilities of 6G systems, such as ultra-

precise 6D localization, sensing, and artificial intelligence functionality, additional
KPIs for these capabilities need to be considered. KPIs for novel capabilities are
discussed in [23].
In addition to KPIs, the social and economic trends towards 2030 motivate addi-
tional indicators for the fulfilment of key values, such as sustainability, inclusiveness,
trustworthiness, and flexibility. To this end, the key value indicators (KVIs) and a
methodology for value representation is introduced and described in [21, 22], and
[23], with further alignment across different projects towards a unified methodol-
ogy happening in the 6G-IA [24].
The main point in this methodology is that the use cases introduced at the begin-
ning of this section can contribute to the key values, and a KVI is used to illustrate
this. When feasible, it is proposed to use the target level of the UN SDGs as a
preferred framework for identifying and detailing the value impact. In some cases,
a KVI quantification may be challenging, and then a connection to a KPI can be
made to grade or assess a value creation potential and contribution from a use case.
For the key value of trustworthiness, a “level of trust” as a KVI is explored, and for
flexibility, an association is made to proxies such as scalability requirements/KPIs.
A KVI analysis of a selected set of use cases is included in [23].

2.2 The Need for a New Architecture

This section presents the overall direction that the 6G architecture should move
towards to fulfil the trends and technology evolutions. This is done by defining a
set of architectural principles and followed by a high-level E2E architecture view.
22 Architecture Landscape

Exposure of AI for full

capabilities automation

Network
simplification in
Extensibility
comparison to
and flexibility
previous
generations
Architectural
Principles
Separation of
concerns of
Scalability
network
functions

Exposed Resilience
interfaces are and
service based availability

Figure 2.3. 6G architecture principles guiding the architecture design [1].

2.2.1 Architectural Principles

To serve as a guideline when developing the 6G architecture, eight different archi-
tectural principles are defined [24, 26]. The order or the numbering of the princi-
ples does not indicate the level of importance. Figure 2.3 shows a summary of the
eight different principles.

Principle 1: Exposure of capabilities

The architecture solution shall expose new and existing network capabilities to E2E
applications and management, such as predictive orchestration. The analytic informa-
tion can, for example, be performance for predictions, such as latency and throughput,
or it can also be localization and sensing information.

Principle 2: AI for full automation

The architecture should support full automation to manage and optimize the network
without human interaction. The closed-loop network automation assumes the use of
AI/ML.

Principle 3: Extensibility and flexibility

The ability of the network to adapt to various topologies without loss of performance
while still enabling easy deployment. This can, for example, be the ability to adapt to
new traffic demands, spectrum situations, private networks, and ad-hoc mesh networks.
The Need for a New Architecture 23

Principle 4: Scalability
The network architecture needs to be scalable both in terms of supporting very small to
very large-scale deployments, by scaling up and down network resources based on needs.

Principle 5: Resilience and availability

The architecture shall be resilient in terms of service and infrastructure provisioning
using features, such as multi-connectivity and removing single points of failure.

Principle 6: Exposed interfaces are service based

Network interfaces should be designed to be cloud-native, utilizing state-of-the-art cloud
platforms and IT tools in a coherent and consistent manner.

Principle 7: Separation of concerns of network functions

The network functions have a bounded context, and all dependencies among services are
through their application programming interfaces (APIs) with a minimal dependency
with other network functions, so that network functions can be developed, deployed, and
replaced independently from each other.

Principle 8: Network simplification in comparison to previous generations

The network architecture should be streamlined to reduce complexity by utilizing cloud-
native upper-layer RAN and CN functions with fewer (well-motivated) parameters to
configure and fewer external interfaces.

2.2.2 End-to-end Architecture

Figure 2.4 depicts a high-level view of the envisioned 6G architecture and highlights
the key technical enablers. The various building blocks are organized into three
layers: Infrastructure, Network Service (NS), and Application.
The infrastructure layer comprises RAN (addressed more in Chapter 3), CN,
and transport networks, which contain radio equipment, switches, routers, com-
munication links, data centres, cloud infrastructure, and so on. The infrastructure
layer provides the physical resources to host the NS and application layer elements.
The envisioned 6G infrastructure layer should also contain RAN improvements,
such as extremely low latency, high reliability, high availability, high data rate, high
capacity, affordable coverage, and high energy efficiency. Extremely high data rate
links will be required in some very high-performance applications anticipated in
6G, e.g., immersive smart cities (a use case from telepresence use case family)
and fully merged cyber-physical worlds (a use case from twinning use case fam-
ily) (see Section 2.1). Most of those are related to highly advanced online imaging,
including holographic communications as well as providing extreme data rates for
24 Architecture Landscape

high-capacity cells. In those cases, a throughput of 100 Gbit/s or even significantly

higher can be required. This means bandwidths of several tens of GHz would be
required to provide this. The architecture design, in particular the infrastructure
layer, needs to ensure that such data rates can be brought to local small-scale base
stations that will serve end users. More details can be found in Chapter 3.
Furthermore, due to the introduction of new use cases and their strict require-
ments, e.g., immersive smart city [22], the infrastructure layer envisioned for 6G
should be able to accommodate new enablers, such as localization and sensing
(addressed more in Chapter 4). Joint communication and sensing (JCAS), also
known as integrated sensing and communication (ISAC), will be one of the main
differentiators of the vision for 6G architecture with respect to 5G communication
systems. Sensing not only includes positioning but also encompasses other novel
functionalities that were not present in 5G, such as radar-type sensing and non-
radar-type sensing using communication technologies, which in turn leads to new
services, such as sensing as a service (SaaS), and landscape sensing [23].
The deployment of mobile networks has become increasingly complex and
diverse with every new generation. The 6G network of networks should easily and
flexibly adapt to new topologies to meet the requirements of both extreme perfor-
mance and global service coverage well beyond what 5G is capable of. The 6G archi-
tecture incorporates different (sub)network solutions into a network of networks.
The 6G network should also be able to support very small to very large-scale deploy-
ments, by scaling up and down network resources based on needs (see Chapter 5).
4G brought the so-called heterogeneous network (HetNet) solutions, i.e., how net-
works with both wide-area macro- and small-cell pico-base stations should cooper-
ate. The extension of the radio spectrum into mmWave in 5G added yet another
aspect to flexible deployment. 6G deployments will include nodes using even higher
sub-THz spectrum (e.g., in the 100–300 GHz frequency range) with limited cov-
erage as well as nodes at low frequencies with seamless coverage. Furthermore, the
number of network solutions for capacity and coverage is also expected to increase
in the 6G timeframe. These include solutions such as distributed multiple-input
multiple-output (D-MIMO) networks, non-terrestrial networks (NTN), campus
networks, mesh networks, and cloudification of the network elements. Thus, 6G
will be a network of networks.
Even with the new 6G solutions mentioned, the increased use of mobile broad-
band and digital solutions may require a more densified network, in order to cope
with the increased capacity needs. This could lead to an increase in overall emissions
unless energy efficiency continues to be addressed.
The envisioned 6G architecture will employ a number of key sustainability
enablers to complement the 6G sustainability targets; it is fundamental to jointly
take into account all the sustainability aspects of networking, including hardware,
The Need for a New Architecture 25

planning, deployment, operations, and the entire equipment life cycle. These
aspects can be effective in achieving sustainability in all layers and levels of the archi-
tecture, namely at deployment levels that include architectural and hardware inno-
vations, at management and orchestration levels that target network operation
efficiency maximization, at service/application layers, such as application-aware
networks, and at cross-layer sustainability enablers that include innovations
in two or more layers. Detailed information on these enablers can be found in
Chapter 6.
The NS layer is envisioned to be cloud- and micro-service-based with functions
and microservices expanded from central cloud to the edge cloud (see Figure 2.4).
One of the key technology enablers of the NS layer is the introduction of the
extreme edge cloud (see Figure 2.4)). Extreme edge cloud covers part of the net-
work with high heterogeneity of devices with a wide variety of technologies, in
terms of both hardware and software. These devices could be personal devices
(smartphones, laptops, etc.) and a huge variety of IoT devices (wearables, sensor
networks, connected cars, industrial devices, connected home appliances, etc.).
The concepts of edge and extreme edge computing become more and more rel-
evant for the 6G architecture and services. Microservice-based implementation can
provide improvements towards a softwarized, intelligent, and efficient 6G archi-
tecture. Chapter 5 describes the enablers for an intelligent network. The ultimate
target for the 6G architecture is to enable autonomous and adaptable networks,
with no (or minimal) human intervention, leveraging cognitive, closed-loop con-
trol network functions that can be instantiated on an on-demand basis, even across
network domain boundaries. In this sense, an intelligent 6G architecture should
be able to define the underlying mechanisms to support embedded AI for 6G and
ensure dynamic adaptability of the network architecture to new use cases while
keeping the infrastructure and energy costs at acceptable and sustainable levels.
Another important aspect of a more flexible and intelligent network is pro-
grammability, addressed in detail in Chapter 7. Programmability can be a tool to
introduce new features, especially to deployments that have a limited footprint due
to limited hardware types and specific requirements. Over the last decade, pro-
grammability is significantly enhanced thanks to the software-defined networking
(SDN) paradigm as well as the ongoing trend towards softwarization and cloudi-
fication. For 6G architecture, this trend is expected to continue in order to allow
third-party developers to interact with the network in new ways, and 6G architec-
ture is expected to ensure reusability and flexibility.
Furthermore, with a cloud-native approach, the RAN and CN architectures
can be streamlined, e.g., reduce some complexity by removing multiple process-
ing points for a certain message and removing duplication of functionalities among
functions [30].
 26

Application Instance #N Fully merged cyber-physical worlds

OSS Application Instance #2 AI-assisted Vehicle-to-Everything (V2X)
2X
X))
X Security
Application Instance #1 Immersive smart cities
Distributed
Ledger
Technologies

AI/ML
Security
Exposure Framework and Integraon Fabric
Quantum
Security

Privacy
Extreme/Far Edge Edge Cloud Central Cloud Enhancing
Technologies

Physical

Management and Orchestration

Layer

Artificial Intelligence and Machine Learning

Security
Trust
CI/CD Foundations

Cloud infrastructure

Localisation Communication Public/Private cloud Transport node

Sensing Function/Micro service Edge node

Figure 2.4 End-to-end system view of the 6G architecture [23].

Architecture Landscape
The Need for a New Architecture 27

Cloud-native technologies can enable the creation of cloudlets at the edge of

the network, with application-to-application and function-to-function communi-
cations, which are capable to satisfy a large number of interconnected assets with
flexible mesh topologies. Another important aspect of the NS layer is the expo-
sure framework and integration fabric. It establishes a communication channel that
enables seamless interoperation and networking across different domains.
The 6G architecture will be able to support the strict requirements of various
use cases that have been envisioned for the next generation of mobile networks (see
Section 2.1.3). In particular, use cases belong to massive twining and telepresence
use-case family, e.g., immersive smart city [22] that can be a digital replica of a
real traffic scenario of the city, the automated train operations, the control of the
utilities (energy, water, gas, etc.), air quality and more are some of the aspects of the
implementation of massive twining to city environments. An interactive 4D map
can be used to plan utility management, such as public transport, garbage, piping,
cabling, buildings, and heating, or to connect many parts of a factory that can be
inspected and steered in detail. Similarly, AI-assisted vehicle to everything (V2X) is
another example of use cases that can provide high level of safety and security for
any transport system, especially road transport due to the prevalence of accidents.
This motivates the need to further explore the potentiality of the AI algorithms for
enhanced automotive services provided by future 6G networks, and it requires a
solid architectural foundation.
Network M&O is gradually moving towards increasing the levels of automation
and fully automated closed-loop control. This is supported by the parallel adop-
tion of advancements in AI/ML technologies. The aim of this shift is to provide a
framework to optimally support reliability, flexibility, resilience, and availability and
addressing changes in the infrastructure, requirements, and failures. More details
on the 6G M&O architecture envisioned for 6G can be found in Section 2.4.
Security and privacy mechanisms are integral parts of the overall architecture,
affecting all network layers as well as the M&O domain. Figure 2.6 highlights the
6G security technology enablers across different layers [31].
Privacy-enhancing technologies are important on all layers where sensitive data
are gathered or processed, and clearly also in the management domain. Similarly,
AI/ML security is relevant for all functions making use of AI/ML, in the sense
of specifically protecting this use, but also refers to AI/ML-driven security mecha-
nisms, e.g., in the management domain [32]. Finally, distributed ledger technolo-
gies are relevant wherever it is required to establish “distributed trust,” i.e., trust that
is not anchored in a centrally trusted authority, as it may be the case in inter-domain
management.
Figure 2.5 shows one possible functional view of the envisioned 6G architecture,
which is depicted on the E2E system view of the architecture. It is hierarchically
 28

Application Instance #N Fully merged cyber-physical worlds

OSS/BSS Application Instance #2 AI-assisted Vehicle-to-Everything (V2X) Security
Application Instance #1 Immersive smart cities
Distributed
Ledger
Technologies

Exposure layer AI/ML

Security
Exposure Framework and Integration Fabric

Quantum
Security
Control Plane Network Functions

Security & privacy Stratum

Privacy
Extreme/Far Edge Edge Cloud Central Cloud Enhancing
Technologies

Sensing Stratum
User Plane Network Functions Network Stratum

Intelligence Stratum
Physical

Management and Orchestration

Layer

Artificial Intelligence and Machine Learning

Security
Trust
CI/CD Foundations

Cloud infrastructure

Localisation Communication Public/Private cloud Transport node

Sensing Function/Micro service Edge node

Figure 2.5 Functional view of the proposed 6G reference architecture with the focus on the stratums.
Architecture Landscape
The Need for a New Architecture 29

composed of the set of planes that traditionally build the mobile network
architecture and has done so since the earliest releases of the 3rd generation part-
nership project (3GPP) standards. In this context, and by borrowing and extending
the terminology from the 3GPP system, a stratum is defined as a set of coordinated
functions that is running in different planes or domains of the network. For the
proposed functional architecture, four stratums on Network, Sensing, Security &
Privacy, and Intelligence have been introduced.
Network stratum is consisted of network functions in control plan (CP) and
user plan (UP) that are responsible for delivering the expected QoS, efficiently
allowing user equipment (UE) to exchange data with the network. CP and UP
entail novel access technologies, which may also include the ones leveraging sub-
terahertz bands and visible light communications; AI-native air interface, arranged
in specific ways (e.g., cell free networks [33, 34]), and even including extreme
edge functions like the ones that are managing and reconfiguring intelligent meta-
surfaces.
Traditionally, the non-access stratum included functions from the UE, UP, and
CP. The network intelligence stratum encompasses and coordinates functions
in all networks, ranging from the intelligent operation of network functions to
their autonomous management and orchestration. The network intelligence stra-
tum gathers data and analytics from the infrastructure layer.
The infrastructure can be extended to include environmental aspects (i.e., the
environment where the infrastructure is deployed, and functions are executed) to
allow a tight interaction between the network and the surrounding space. Properly
steering beams at very high frequencies or using unmanned aerial vehicle (UAV)
to extend the network’s coverage requires a sensing stratum that can efficiently
coordinate functions, harvesting data from fixed landmarks or dynamic laser/light
imaging, detection, and ranging (LIDAR) scans, or even using the UP wireless
technology as an additional source of sensing, possibly in an energy harvesting
fashion.
The last stratum is the security & privacy stratum, which manages the cyber
security and data privacy aspects of the network. This stratum coordinates functions
in all the planes and domains of the network up to the vertical service provider
one, which also benefits from the enhanced 6G security and cooperates with it
to minimize the attack surface, while allowing the service customers to have full
control over the data (including the network one).
Clearly, this richness in the available network functions, which have to be
arranged and properly configured according to the network slices they belong to,
poses new challenges to the management plane of the network, see Section 2.4.
This interaction is possible thanks to the enhanced exposure interface between
the network and the vertical service providers on the application layer, through the
30 Architecture Landscape

use of network applications, which can leverage on data, functions, and procedures
offered to support and enhance the user experience. Through the exposure inter-
face, the traditional barrier between operators and service providers is removed,
allowing a white-box customization of the vertical services.

2.3 Security & Privacy Architectural Components

Figure 2.6 shows the overall architecture, visualizing the applicable security and pri-
vacy components in all areas, and highlighting the specific 6G security technology
enablers. While the focus lies on the technology enablers as new architectural com-
ponents, a holistic 6G security architecture must also comprise today’s well-proven
security mechanisms, as far as they are still relevant in 6G. On this basis, Figure 2.6
summarizes these components, without the aspiration of exhaustiveness and depth
of detail.
Figure 2.6 distinguishes among non-virtualized equipment (for radio access and
optical transport), the cloud infrastructure, and the software running on it, includ-
ing the virtualization layer, the logical network layer, and the management and
orchestration functions, including security and risk management and inter-domain
management. In each part, the figure shows the most relevant security and privacy
building blocks or architectural components, with the new 6G security technol-
ogy enablers highlighted in red, and the more traditional building blocks, like for
example “Secure SW,” in blue.
Many building blocks apply to multiple areas, e.g., “Secure SW ” applies to the
non-virtualized radio and optical equipment (as far as this equipment comprises
software), to the virtualization layer, and to all the software running on it, including
M&O functions. As another example, “Trust foundations” apply to all hardware,
i.e., the radio and optical equipment as well as the cloud infrastructure. On the
other hand, some building blocks appear in dedicated places only, like “Distributed
ledger technologies” appearing at inter-domain management only, but this does not
preclude the potential applicability of the building block in other areas. Also, when
a building block appears in an area, this does not imply that the building block
is always applicable. For example, certain non-virtualized radio access equipment
may not have access to sensitive data, so no privacy-enhancing technologies may
be required here. As another example, obviously not all transport equipment is
required to support quantum key distribution.
The traditional security building blocks may be mostly self-explaining, but note
the following:
• “Secure SW” refers to software with a low (close to zero) degree of vulnera-
bility. “Secure HW/FW” has the same meaning for hardware or firmware. An
 Secure SW Secure protocol & API design Cloud-based Software
Secure AI/ML Privacy enhancing technologies Quantum-safe cryptography

Interdomain Application Instance #n Application Layer

Management Application Instance #2
Distributed
ledger Application Instance #1
technologies

Security and Risk

Management
Virtualisation Layer Virtualised Network Functions
Network Service Layer
Infrastructure security functions Secure logical network topology
AI/ML-driven
(security isolation, resource (slices, security zones, traffic
automated
isolation, filtering functions) separation)
closed loop
security
mechanisms

Secure SW Quantum-safe cryptography Non-Virtualized Equipment

Security & Privacy Architectural Components

“Classical”
management RAN Equipment Privacy enhancing Optical Equipment
security
technologies
mechanisms Physical layer security Quantum key Infrastructure Layer
Jamming protection Secure AI/ML distribution

Management and Orchestration

Attestation
technologies Cloud
Crypto accelerators Secure HW/FW Trust foundations
Infrastructure
Figure 2.6 Overview of the essential 6G security architectural components [33]. The new 6G security technology enablers highlighted in red,
and the more traditional building blocks are in blue.
31
32 Architecture Landscape

example is the robustness of a processor against leaking information between

different processes running on this processor in a (quasi-) parallel manner.
• “Secure protocol and API design” refers to robustness not only against external
attackers (which is typically achieved by the use of cryptography), but also
against erroneous or malicious behaviour of authorized peers.
• “Classical management security mechanisms” comprise well-established mech-
anisms, such as access control, role-based access, secure logging, isolation of
management functions/traffic from all other traffic, etc.
Further details on the components of the security and privacy-preserving tech-
nologies are provided in Chapter 8.

2.4 Service Management and Orchestration

Service M&O deals with the deployment and operation of the NSs supplied
through the mobile network operator (MNO) to their customers, preserving all of
the contractual aspects associated to those services. It addresses the provision of ser-
vices, QoS and quality of experience (QoE) fulfilment, or fault reporting, among
others. In previous generations of the mobile communications systems, the cus-
tomers of the MNOs have been mainly individuals consuming voice and messaging
services. However, the market situation is much more complex now, including new
data services and corporate customers, such as vertical industries, digital operators,
hyper-scalers, or large-scale content providers, among others. It is anticipated that
this trend, in terms of heterogeneity of stakeholders and provided services, could
continue and even experience growth within the coming years.
To cope with this complexity, it is needed to enable the services M&O sys-
tems with the required capabilities to provide the necessary orchestration resources.
Specifically, the following main capabilities have been identified for the future 6G
M&O systems.
The adoption of the cloud-native principles also in the M&O system.
This would be aligned with the E2E architectural concepts in Section 2.2.2, but
from the M&O perspective, it would involve three main aspects: (i) the prior-
ity on using micro-services, i.e., light-weight self-contained, independent, and
reusable components from different suppliers; (ii) the implementation of the service
mesh concept, regarding the communication among the network components; and
(iii) the enabling mechanisms for the NSs to be deployed/updated using “contin-
uous” DevOps-like practises, e.g., implementing CI/CD workflows with a high
automation degree.
Unified M&O across multiple domains that could be owned/administered
by multiple stakeholders and featured with heterogeneous technology resources.
This entails the definition of converging interfaces, the mechanisms to dynamically
Service Management and Orchestration 33

check and expose the different resources and capabilities from each domain, and
the access control procedures for consuming the various primitives and services.
Increased degree of automation to strongly reduce manual interventions
regarding the functionalities of service and network planning, design, provision-
ing, optimization, and operation/control, leveraging closed-loop and zero-touch
responses. The M&O system needs to be able to identify, detect, or predict poten-
tial issues, triggering automatic reactions.
Adoption of data-driven and AI/ML techniques in the M&O system. AI/ML
techniques could cover numerous optimization aspects and lifecycle actions con-
cerning the services M&O, including resource allocation and slice sharing at pro-
visioning time, service composition, scaling, migration, re-configuration, and re-
optimization of NSs, among others.
Intent-based approaches for service planning and definition. In order to
help with the extended complexity, the M&O system would implement automated
mechanisms for translating service specifications and commands based on high-
level intents, which might be expressed even in natural language (e.g., relying on
AI/ML techniques).
To meet these main challenges, the M&O system is seen as a common function-
ality impacting all layers of the E2E architecture: from the infrastructure up to the
applications (see Figure 2.7). In this regard, an initial high-level M&O architectural
design for the future 6G networks has been produced. This architectural design
takes the previous 5G architectural view from the 5G PPP Architecture Working
Group as a baseline [34, 35] represents the structural view of this architecture, with
the main building blocks grouped in different layers.
The NSs and slices at the service layer (top in Figure 2.7) are executed on the
infrastructure layer (bottom) through the network functions at the network ser-
vice layer (middle). All these elements (network functions, services, and slices) are
designed and provided from the design layer (right).
A new layer, named the design layer, has been included to represent the M&O-
related operations involving third-party software providers. This is intended to
introduce the well-known DevOps-like practises (e.g., continuous integration and
continuous delivery/continuous deployment (CI/CD)) in the telco-grade environ-
ment. Also, hyperscalers, private networks, and the extreme edge domain have
been explicitly included as part of the infrastructure layer. New control loops have
been included: (i) the “DevOps control loop,” representing the automated con-
tinuous iterations (e.g., CI/CD) between the MNO scope (grey colour) and the
external design layer (light blue colour); and (ii) the “infrastructure control loop,”
meant to automate the infrastructure discovery processes and the related monitor-
ing methods targeting the extreme edge asset integration (which can be potentially
asynchronous in terms of connection/disconnection of devices, so requiring special
 34

M&O Scope
Service Layer (for Vercals) – E2E Slice specific realizaon Design Layer
• Service creaon and
Slice Instance #1 Service D operaon (SA & Service
Implemented though NetApps and
Service B fulfillment control Layer
User Plane, Access, Mobility loops)
Service A Control Service Layer
Service C Management and Session • Intent-based service Design
Management Funcons management Loops M&O
System
Slice Instance #2 • Service Quality Mgmt. Security
• Data aggregaon
Security
Security SW &
Slice Instance #n
Descriptors
Design
Network Layer Network Layer DevOps /
Radio Access Transport Net. Security Management M&O
Made AIOps
Funcons Funcons Funcons Use Funcons of Security Framework
AI AI AI
Security
Security
Primarily
implemented Intent-based
AI AI Network service
through CNFs.
Layer definions
Also, through Use Use Control
VNFs, PNFs or
API Management Exposure

Loops
other NFs Core Network Third Party AI/ML Monitoring
implementao Funcons Funcons Funcons Funcons • NFs LCM
AI AI AI AI
n technologies • Monitoring DevOps
• AI/ML orch. Control
AI AI
acons
• Security policies
Loops
SW

• AI Models
• Predicon
• Data
Infrastructure Layer analycs
Edge • AI Federaon
cloud Infra. Layer Infrastructure
Control Layer M&O
Loops Security
Security
Security • API registry
Extreme-edge Transport Core/Central • API discovery
Cloud • Infrastructure
• Access
discovery
Control
• Monitoring
External
Public Networks Private Networks Infrastructure

Figure 2.7 Proposed 6G management & orchestration system – structural view [34].
Architecture Landscape
Summary and Outlook 35

processes for their management). As in the baseline architecture in [35], NFs are
associated in different groups at the network layer (e.g., radio access functions,
CN functions, M&O functions, AI/ML functions, etc.). However, following the
cloud-native practises, these functions would be primarily implemented through
containerized NFs (CNFs), although also through virtualized NFs (VNFs), phys-
ical NFs (PNFs), or other NF implementation technologies (e.g., to ensure back-
ward compatibility). It should be noticed here that, although some functions work
only as managed resources (e.g., CN functions or third-party functions), others are
specific M&O resources (e.g., the monitoring functions or the management func-
tions themselves); however, other functions are hybrid : they can support M&O
resources (e.g., certain security-related or AI/ML functions) or work as pure man-
aged resources (e.g., certain AI as a Service (AIaaS) functions or security functions
not in the M&O scope). Functions in the network layer are generic, i.e., instead
of referring specific functions (e.g., communication service management function
(CSMF), media resource function (MRF), NFV orchestrator (NFVO), etc.) as
in [35], just generic blocks are provided. This is intentional, in order to consider
the new functions that would be probably defined for the future 6G stack. A new
set of AI/ML collaborative components have been distributed across the network
covering both managing and managed scopes. M&O functions can be instantiated
in the three different layers (service, infrastructure, and network layers), including
specific security-related functions. Finally, and also aligned with the cloud-native
approach, a new cross-layer API management exposure block has been included
to communicate the different network elements in the different network layers. In
short, it mimics the behaviour of the zero-touch service management (ZSM) cross-
domain integration fabric, enabling the so-called capabilities exposure of the network
of elements in the various architectural layers. It makes possible communicating
the various M&O resources within and between administrative domains, although
it could be applied more broadly to represent potential federated interactions.

2.5 Summary and Outlook

This chapter discusses the current architectural trends and technologies for the
future 6G network. Motivated by the surge of new requirements stemming from
societal trends and use cases, a set of architectural principles has been introduced,
and new architectural and technical enablers needed for the 6G architecture have
been identified. A high-level view of a possible E2E system of the 6G architecture as
well as a functional view is described. Thereafter, a description on how the enablers
fit into the system view is given, which is also an overview of the content in this
book. The chapter dives into the security and privacy area in a bit more detail and
36 Architecture Landscape

gives an overview of the 6G security and privacy architectural components. Finally,

the main capabilities needed for a future 6G M&O systems are discussed.

References

[1] United Nations (UN), “Transforming our world: the 2030 Agenda for Sus-
tainable Development,” 21 October 2015. Accessed: April 6, 2023. [Online].
Available: https://sdgs.un.org/2030agenda.
[2] Ericsson, “6G – Connecting a cyber-physical world,” February 2022.
Accessed: April 6, 2023. [Online]. Available: https://www.ericsson.com/4
927de/assets/local/reports-papers/white-papers/6g--connecting-a-cyber-ph
ysical-world.pdf .
[3] Nokia, “Communications in the 6G era,” 9 September 2020. Accessed: April
6, 2023 [Online]. Available: https://onestore.nokia.com/asset/207766.
[4] GSM Association, “2022 Mobile Industry Impact Report: Sustainable Devel-
opment Goals,” September 2022. Accessed: April 6, 2023. [Online]. Avail-
able: https://www.gsma.com/betterf uture/wp-content/uploads/2022/11/2
022-SDG-Impact-Report.pdf .
[5] M. Matinmikko-Blue, S. Aalto, M. I. Asghar, H. Berndt, Y. Chen, S. Dixit,
R. Jurva, P. Karppinen, M. Kekkonen, M. Kinnula, and P. Kostakos, White
paper on 6G drivers and the UN SDGs, University of Oulu, June 2020.
Accessed: April 6, 2023. [Online]. Available: http://urn.fi/urn:isbn:9789
526226699.
[6] G. Fettweis, “6G – Just a better 5G?: 2020 5G World Forum keynote series,”
18 November 2020. Accessed: April 6, 2023. [Online]. Available: https://ieee
tv.ieee.org/2020-5g-world-forum-keynote-gerhard-fettweis.
[7] N. Demassieux, “6G. Why?,” In EuCNC 2020, Dubrovnik, Croatia, 2020.
Accessed: April 6, 2023. [Online]. Available: https://www.eucnc.eu/wp-cont
ent/uploads/2020/07/2020-05-29-6G.-Why-Nicolas-Demassieux-EUCN
C-VDEF.pdf .
[8] Samsung, “The Next Hyber-Connected Experience for All,” 14 July 2020.
Accessed: April 6, 2023. [Online]. Available: https://cdn.codeground.org/nsr
/downloads/researchareas/20201201_6G_Vision_web.pdf .
[9] NTT DOCOMO, “White paper 5G Evolution and 6G, version 5.0,” January
2023. Accessed: April 6, 2023. [Online]. Available: https://www.docomo.ne.
jp/english/binary/pdf/corporate/technology/whitepaper_6g/DOCOMO_
6G_White_PaperEN_v5.0.pdf .
[10] Raconteur, “The economic impact of 5G,” 2020. Accessed: April 6, 2023.
[Online]. Available: https://www.raconteur.net/infographics/the-economic-
impact-of-5g/.
References 37

[11] European Commission, “A New Industrial Strategy for Europe,” 10 March

2020. Accessed: April 6, 2023. [Online]. Available: https://eur-lex.europa.e
u/legal-content/EN/TXT/?uri=CELEX:52020DC0102.
[12] Arcep, “Networks and the Environment,” 8 September 2020. Accessed: April
6, 2023. [Online]. Available: https://en.arcep.f r/fileadmin/cru-1677573101/
user_upload/37-20-english-version.pdf .
[13] The Connexion, “Lille issues ‘moratorium’ on 5G technology,” 11 October
2020. Accessed: April 6, 2023. [Online]. Available: https://www.connexionf
rance.com/article/French-news/Lille-issues-moratorium-on-controversial-5
G-technology-pending-2021-Anses-report.
[14] International Commission on Non-Ionizing Radiation Protection, 2023.
Accessed: April 6, 2023. [Online]. Available: https://www.icnirp.org/.
[15] V. Ziegler, H. Viswanathan, H. Flinck, M. Hoffmann, V. Räisänen, and
K. Hätönen, “6G architecture to connect the worlds,” In IEEE Access, vol. 8,
pp. 173508–173520. Sep. 2020.
[16] Orange, “Orange inaugura tienda en el metaverso,” 13 September 2022.
Accessed: April 6, 2023. [Online]. Available: https://blog.orange.es/innov
acion/orange-inaugura-tienda-en-el-metaverso/.
[17] R. Minerva, G. M. Lee, and N. Crespi, “Digital twin in the IoT context:
A survey on technical features, scenarios, and architectural models,” In Pro-
ceedings of the IEEE, vol. 108, no. 10, pp. 1785–1824, 2020.
[18] European Telecommunications Standards Institute (ETSI), “About ETSI,”
2023. Accessed: April 6, 2023. [Online]. Available: https://www.etsi.org/a
bout.
[19] M. Rasti, S.K. Taskou, H. Tabassum, and E. Hossain, “Evolution toward 6g
multi-band wireless networks: A resource management perspective,“ In IEEE
Wireless Communications, vol. 29, no. 4, pp. 118–125.
[20] Z. Zhang, Y. Xiao, Z. Ma, M. Xiao, Z. Ding, X. Lei, G. K. Karagiannidis,
and P. Fan P, “6G wireless networks: Vision, requirements, architecture, and
key technologies,” In IEEE Vehicular Technology Magazine, vol. 14, no. 3,
pp. 28–41, Jul 2019.
[21] Hexa-X, “D1.1 – 6G Vision, use cases and key societal values,” 28 February
2021. Accessed: April 6, 2023. [Online]. Available: https://ec.europa.eu/res
earch/participants/documents/downloadPublic?documentIds=080166e5d9f
611f 1&appId=PPGMS.
[22] Hexa-X, “D1.2 – Expanded 6G vision, use cases and societal values – includ-
ing aspects of sustainability, security and spectrum,” 30 April 2020. Accessed:
April 6, 2023. [Online]. Available: https://ec.europa.eu/research/participants
/documents/downloadPublic?documentIds=080166e5dc8b611b&appId=P
PGMS.
38 Architecture Landscape

[23] Hexa-X, “D1.3 – Targets and requirements for 6G – initial E2E architecture,”
28 February 2022. Accessed: April 6, 2023. [Online]. Available: https://ec.e
uropa.eu/research/participants/documents/downloadPublic?documentIds=0
80166e5e90431aa&appId=PPGMS.
[24] 6G Infrastructure Association, Vision and Societal Challenges Working
Group, Societal Needs and Value Creation Sub-Group, “What societal values
will 6G address?,” May 2022. Accessed: April 6, 2023. [Online]. Available:
https://5g-ppp.eu/wp-content/uploads/2022/05/What-societal-values-will
-6G-address-White-Paper-v1.0-f inal.pdf .
[25] Hexa-X, “D5.1 – Initial 6G architectural components and enablers,” Decem-
ber 2021, Accessed: April 6, 2023. [Online]. Available: https://ec.europa.eu/r
esearch/participants/documents/downloadPublic?documentIds=080166e5e
652c1f 2&appId=PPGMS.
[26] M. Ericson, S. Wänstedt, M. Saimler, H. Flinck, G. Kunzmann, P. Vlacheas,
P. Demestichas, D. Rapone, A. De La Oliva, C. J. Bernardos, and R. Bassoli,
“Setting 6g architecture in motion–the hexa-x approach,” In 2022 Joint Euro-
pean Conference on Networks and Communications & 6G Summit (EuCNC/6G
Summit), June 2022.
[27] Hexa-X, “D3.2 – Localisation and sensing use cases and gap analysis,” Septem-
ber 2022. Accessed: April 6, 2023. [Online]. Available: https://ec.europa.eu/r
esearch/participants/documents/downloadPublic?documentIds=080166e5e
648d02a&appId=PPGMS.
[28] Hexa-X Deliverable “D5.2 – Analysis of 6G architectural enablers’ applica-
bility and initial technological solutions,” October 2022, Accessed: April 6,
2023. [Online]. Available: to appear in https://cordis.europa.eu/project/id/
101015956/results.
[29] Hexa-X Deliverable “D2.1 – Towards Tbps Communications in 6G: Use
Cases and Gap Analysis” June 2021. Accessed: April 6, 2023. [Online]. Avail-
able: https://ec.europa.eu/research/participants/documents/downloadPubli
c?documentIds=080166e5df2529f d&appId=PPGMS.
[30] E. U. Soykan, L. Karaçay, F. Karakoç, and E. Tomur, “A Survey and Guideline
on Privacy Enhancing Technologies for Collaborative Machine Learning,” In
IEEE Access, vol. 10, pp. 97495–97519, Sep 2022.
[31] N. Rajatheva, I. Atzeni, E. Björnson, A. Bourdoux, S. Buzzi, J. B. Dore,
S. Erkucuk, M. Fuentes, K. Guan, Y. Hu, and X. Huang, J. Hulkkonen,
J. M. Jornet, M. Katz, R. Nilsson, E. Panayirci, K. Rabie, N. Rajapaksha,
M. J. Salehi, H. Sarieddeen, T. Svensson, O. Tervo, A. Tolli, Q. Wu, and
W. Xu, “White paper on broadband connectivity in 6G,” In arXiv preprint
arXiv:2004.14247. April 2020.
References 39

[32] H. Q. Ngo, A. Ashikhmin, H. Yang, E. G. Larsson, and T. L. Marzetta, “Cell-

free massive MIMO versus small cells,” In IEEE Transactions on Wireless Com-
munications, vol. 16, no. 3, pp. 1834–1850, Jan 2017.
[33] M. B. Khorsandi, R. Bassoli, G. Bernini, M. Ericson, H. F. Fitzek, A. Gati,
H. Harkous, H. Hoffmann, I. L. Pavon, G. Landi, and D. Lopez, “6G E2E
Architecture Framework with Sustainability and Security Considerations,” In
2022 IEEE Globecom Workshops (GC Wkshps) Dec 4, 2022.
[34] Hexa-X Deliverable “D6.2 – Design of service management and orchestration
functionalities”, Apr. 2022. Accessed: April 6, 2023. [Online]. Available: to
appear in https://cordis.europa.eu/project/id/101015956/results.
[35] 5GPPP, “Architecture Working Group – View on 5G Architecture”, Version
4.0, August 2021. Accessed: April 6, 2023. [Online]. Available: https://5g-p
pp.eu/wp-content/uploads/2021/11/Architecture-WP-V4.0-final.pdf .